European

www.ejconline.com

European Journal of Cancer 41 (2005) 807–815

Journal of

Cancer

Human papillomavirus-16 associated squamous cell carcinoma ofthe head and neck (SCCHN): A natural disease model

provides insights into viral carcinogenesis

Robert L. Ferris a,b,*, Ivan Martinez d, Nicky Sirianni b, Jun Wang b,Andres Lopez-Albaitero b, Susanne M. Gollin c, Jonas T. Johnson b, Saleem Khan d

a UPCI Research Pavilion, The Hillman Cancer Center, 5117 Centre Avenue, Room 1.19d, Pittsburgh, PA 15213-1863, USAb Departments of Otolaryngology and Immunology, University of Pittsburgh Medical Center and Cancer Institute, Pittsburgh, PA, USA

c Department of Human Genetics, University of Pittsburgh Graduate School of Public Health and University of Pittsburgh Cancer Institute,

Pittsburgh, PA, USAd Department of Molecular Genetics and Biochemistry, University of Pittsburgh School of Medicine, Pittsburgh, PA, USA

Received 11 October 2004; accepted 23 November 2004

Abstract

Uncertainty regarding the causality of human papillomaviruses (HPVs) in squamous cell carcinoma of the head and neck

(SCCHN) necessitates better in vitro models. We carried out molecular analyses of a novel, naturally HPV-16-transformed SCCHN

cell line (UPCI:SCC090) and show high copy number of HPV-16 DNA, present in a head to tail, tandemly repeated integrated state.

Sequence analysis of the HPV-16 long control region (LCR) in UPCI:SCC090 revealed a deletion of 163 bp, removing a portion of

the enhancer sequence, including the binding sites for the transcription factors YY1 and NF1. The E6 and E7 oncogenes of HPV-16

are expressed at high levels in this cell lines, as determined by quantitative reverse transcriptase-polymerase chain reaction (RT-

PCR). UPCI:SCC090 contains wild-type tumour suppressor TP53 gene, and undetectable p53 protein, except after treatment with

cisplatin, specific proteasome inhibitors or by E6 RNA interference, suggesting E6-dependent degradation of p53 in this cell line.

The results of our studies are consistent with a causative role of HPV-16 in the pathogenesis of SCCHN.

� 2005 Elsevier Ltd. All rights reserved.

Keywords: Human papillomavirus; Head and neck cancer

1. Introduction

Human papillomaviruses (HPVs) are small, double-

stranded DNA viruses that are associated with upper

and lower genital tract neoplasias [1,2]. HPV types 16and 18 are associated with a majority of cases of

HPV-induced cervical cancers [3–5]. In most benign

and preneoplastic cervical lesions, the HPV DNA is

present as a nuclear plasmid. However, in carcinomas

the HPV DNA is usually found to be integrated into

0959-8049/$ - see front matter � 2005 Elsevier Ltd. All rights reserved.

doi:10.1016/j.ejca.2004.11.023

* Corresponding author. Tel.: +1 412 623 1416; fax: +1 412 623 1415.

E-mail address: ferrisrl@upmc.edu (R.L. Ferris).

one or more human chromosomes [6], leading to upreg-

ulation of the viral E6 and E7 genes and tumour pro-

gression. Although there has been an increase in the

number of studies dealing with the role of HPVs in squa-

mous cell carcinoma of the head and neck (SCCHN)during carcinogenesis, such studies have been hampered

by the inability to study the disease in vitro. Such studies

require a well-characterised cellular model of HPV-asso-

ciated SCCHN for molecular analyses similar to those

used in the study of cervical carcinomas [7].

There were approximately 500000 new cases of

SCCHN worldwide in 2001 [8], usually associated with

such risk factors as heavy consumption of alcohol and/

808 R.L. Ferris et al. / European Journal of Cancer 41 (2005) 807–815

or tobacco. Although the distribution of episomal and

integrated HPV forms in both precancerous and cancer-

ous lesions of the head and neck has not been deter-

mined, limited evidence suggests similar mechanisms to

those observed in cervical carcinomas [9–13]. Epidemio-

logical and molecular studies have shown that morethan 90% of cervical tumours exclusively harbour

integrated viral sequences [14]. Molecular studies of

HPV-associated SCCHN are necessary for a better

understanding of the physical state and potential role

of this virus in carcinogenesis, and for the development

of new, more targeted therapeutic strategies.

The carcinogenic mechanism of HPV-induced

SCCHN may differ from that of anogenital cancers.Although well-characterised in vitro cellular model sys-

tems exist for cervical carcinoma, including the CaSki,

HeLa and SiHa cell lines, no clearly defined model for

HPV-associated SCCHN exists, despite occasional re-

ports of HPV DNA in head and neck cancer cell lines.

Furthermore, the SCCHN field is complicated by the

reporting of oral or gingival keratinocyte cultures that

are transiently or stably transfected with genes encodingE6 and E7, without the presence of the other HPV genes

[15]. Thus, the ideal in vitro model for HPV-associated

SCCHN would be a naturally HPV-16-transformed cell

line derived from a de novo tumour from the oropharynx

of a SCCHN patient. Here, we report the molecular

characterisation of a recently identified HPV-16+ oro-

pharyngeal cell line (UPCI:SCC090).

2. Materials and methods

2.1. Cell lines

Cell lines were cultured as described in Ref. [16].

SiHa, C-33A, PCI-30, PCI-13 (gifts from Dr. Theresa

Whiteside, UPCI), L-18 ([17], a gift from Lou Laimins),and UPCI:SCC090: DMEM + 10% FBS (fetal bovine

serum) + 2% L-glutamine + 1% Penicillin/Streptomycin

(Invitrogen). For CaSki cells, Roswell Park Memorial

Institute (RPMI) 1640 was used.

The UPCI:SCC090 cell line was derived by the ex-

plant method from a 44-year old male (now deceased)

smoker with an oropharyngeal SCCHN arising in the

base of tongue. His tumour was staged as T2N1M0according to the 4th Edition American Joint Committee

on Cancer (AJCC) guidelines, and the histology was

moderately to poorly differentiated invasive squamous

cell carcinoma with basaloid features.

2.2. HPV-positivity, Southern blotting and the

identification of integrated HPV-16 DNA

DNA from the following cell lines was isolated by

phenol–chloroform–isoamyl alcohol (25:24:1) extraction

and ethanol-precipitation [18]: UPCI:SCC90; the cervi-

cal carcinoma cell line, CaSki, containing tandemly inte-

grated copies of HPV-16 DNA; and L-18, a human

keratinocyte cell line containing episomal copies of

HPV-18 DNA [17]. HPV-positivity in the UP-

CI:SCC090 cell line was tested using M09/M11 PCRprimers which amplify an approximately 450-bp con-

served region of the L1 gene of HPVs [19]. HPV-16-spe-

cific polymerase chain reaction (PCR) was done by

amplifying a 477-bp region of the E6 gene of this virus

using 5 0-ATGCACCAAAAGAGAACTGC-3 0 as the

forward primer, and 5 0-TTACAGCTGGGTTTCTC-

TAC-3 0 as the reverse primer. The glyceraldehyde-3-

phosphate dehydrogenase (G3PDH) gene was amplifiedas a control using the forward primer 5 0-

ACCACAGTCCATGCCATCAC-3 0 and the reverse

primer 5 0-TCCACCACCCTGTTGCTGTA-3 0 which

amplify a 556-bp DNA region. PCR for the L1, E6

and G3PDH were performed in a 50 ll volume contain-

ing 50 mM KCl, 10 mM Tris (pH 8.3), 1.5 mM MgCl2,

0.01% gelatin, 200 lM deoxynucleoside triphosphate

(dNTP) mix, 0.4 lM of each primer and 2.5 units ofthe Taq DNA polymerase. The DNA was denatured

at 94 �C for 5 min, followed by 40 PCR amplification cy-

cles that consisted of denaturation (94 �C, 1 min),

annealing (55–60 �C, 1 min) and extension (72 �C, 2

min). An additional extension step of 72 �C for 5 min

was included at the end of the reaction. The PCR-ampli-

fied DNA was analysed by agarose gel electrophoresis

[20]. For Southern blot analysis, 5 lg of CaSki and UP-CI:SCC090 DNA was digested with three different

restriction endonucleases. BglII does not cleave the

HPV-16 genome, BamHI cleaves it once and KpnI

cleaves it at two sites. For the L-18 cell line, the DNA

was treated with BglII which does not cleave the HPV-

18 DNA, EcoRV (which cleaves it once) and BamHI

that cleaves the HPV-18 DNA twice. The digested

DNA was electrophoresed on a 0.7% agarose gel andthe Southern blots were probed with a 32P-labelled plas-

mid containing the complete HPV-16 genome as de-

scribed in Ref. [18]. The blots were subjected to

autoradiography at �80 �C. The identity of HPV-16 in

UPCI:SCC90 was confirmed by PCR amplification of

portions of the E2, E6, E7 and L1 genes followed by

automated DNA sequencing of the amplified products.

The following HPV-16-specific primers were used forthe PCR amplification. E2 (1,228 bp product) forward

primer 5 0-GGAAATCCAGTGTATGAGCTTAATG-

3 0 and reverse primer 5 0-GTAATGTTGTGGATG-

CAGTATCAAG-3 0; E6/E7 (735 bp product) forward

primer 5 0-ATGCACCAAAAGAGAACTGC-3 0 and re-

verse primer 5 0-TGCCCATTAACAGGTCTTCC-3 0.

The MY09/MY11 primers [19] were used for the ampli-

fication of the L1 gene and primers described above forthe amplification of the E6 region of HPV-16. The PCR

conditions were the same as described above, except for

R.L. Ferris et al. / European Journal of Cancer 41 (2005) 807–815 809

the E2 gene for which extension reactions were carried

out for 2.5 min. DNA sequencing was carried out using

an Automated Applied Biosystems PRISM 3100 Genet-

ic Analyzer.

2.3. Identification of the region deleted in the HPV-16

DNA in the UPCI:SCC090 cell line

The long control region (LCR) of the HPV-16 DNA

was amplified by PCR. The sequences of the primers

used were: 5 0-TTTTGGCACAAAATGTGTTTTT-3 0

for the forward primer (HPV-16 positions 7470–7491)

and 5 0-GCACAGAGCTG CAAACAACTAT for the

downstream primer (positions 150–171). The reactionmixtures contained 200 lM of each deoxynucleoside tri-

phosphate (dNTP), 200 ng of UPCI:SCC090 DNA, 1

lM of each primer, and 5 units of the Pfu polymerase

(Stratagene, La Jolla, CA). The conditions of amplifica-

tion were as follows: 94 �C for 5 min; 94 �C for 1 min, 60

�C for 1 min, and 72 �C for 2 min for 40 cycles; and 72

�C for 7 min. The PCR-amplified DNA was isolated by

gel electrophoresis and subjected to automated DNAsequencing.

2.4. Quantitative real-time PCR (qPCR)

Relative HPV-16 E6 and E7 DNA copy number in

the UPCI:SCC090 cells was determined by type-specific

primers/probe and conditions [21]. Control DNA quan-

tification was performed, amplifying a series of micro-satellite repeats (QuMA) [22], and using a serially

diluted HPV-16 E6-encoding plasmid. Comparison

was made with CaSki and SiHa (see Table 1). Input

copy numbers were determined using HPV-16 E6-con-

taining plasmid DNA, and unknown samples norma-

lised to E6 input plasmid amounts. Relative expression

of E6 and E7 was calculated using the delta CT method

described previously in Ref. [22]: (Relative expres-sion = 2�DCT; where DCT = CT(Target gene) � CT(QuMA)).

While this equation is exactly accurate only for quanti-

tative real-time RT-PCR (qRT-PCR) reactions that

are 100% efficient, it provides an estimate, particularly

when compared with internal known reagents, such as

the E6-encoding plasmid and the CaSki cell line, which

Table 1

HPV-16 qPCR in SCC90 cells

Cell line [E6]a [E7]a

SCC90 91 171

CaSki 105 249

SiHa 0.46 1.0

qPCR, quantitative real-time PCR.a Relative to control QuMa DNA amplification, using 2�DCT

method.

has been quantified in terms of the integrated HPV

DNA present.

2.5. Isolation of RNA and RT-PCR analysis

Expression of the HPV-16 E2 and E6 genes wasinvestigated by reverse transcriptase (RT)-PCR analy-

sis. RNA was isolated using the ULTRASPEC RNA

isolation system (Biotecx) according to the manufac-

turer�s protocol. Before DNA synthesis, RNA was

treated with DNaseI, amplification grade (Invitrogen)

for 15 min at room temperature to avoid DNA con-

tamination. DNaseI was then inactivated by the addi-

tion of 25 mM ethylene diamine tetraacetic acid(EDTA) followed by incubation at 65 �C for 10

min. The lack of contaminating DNA was confirmed

by the failure of PCR amplification in reactions con-

taining the Taq polymerase, but lacking the reverse

transcriptase. The cDNA synthesis was performed at

37 �C for 1 h in a final volume of 20 ll using 1 lgof total RNA template, 0.5 lg of oligo (dT)15, 10

mM dNTPs, 30 U of RNase inhibitor and 200 unitsof MMLV reverse transcriptase. Expression of the

HPV-16 E6 gene was studied using the primer pair

described above. To detect expression of the HPV-16

E2 gene, PCR amplification was done using 5 0-

AAAGTGGACATTACAAGACGTTAGC-3 0 for the

forward primer and 5 0-GTGAGCTGTTAAATG-

CAGTGAGG-3 0 for the reverse primer that are ex-

pected to generate a 554-bp product. The expressionof the cellular G3PDH gene was used as a control.

PCR was performed in a 50 ll volume containing

50 mM KCl, 10 mM Tris (pH 8.3), 1.5 mM MgCl2,

0.01% gelatin, 200 lM dNTP mix, 0.4 lM of each pri-

mer and 2.5 units of the Taq DNA polymerase. The

DNA was denatured at 94 �C for 5 min, followed

by 40 PCR amplification cycles that consisted of dena-

turation (94 �C, 1 min), annealing (55–60 �C, 1 min)and extension (72 �C, 2 min). An additional extension

step of 72 �C for 5 min was included at the end of the

reaction. The PCR products were analysed by electro-

phoresis on 1% agarose gels.

2.6. Quantitative real-time RT-PCR

Reverse transcription was performed with randomhexamer primers and Superscript � (Invitrogen Corp.)

as described previously in Ref. [9]. As described in

Ref. [21], qRT-PCR was then carried out on the Applied

Biosystems 7700 Sequence Detection Instrument at 95

�C for 12 min, PCR was performed at 95 �C for 15 s,

60 �C for 60 s. Relative expression of the target gene:

endogenous control gene, b-glucuronidase (GUS), was

calculated using the delta CT method described previ-ously: (Relative expression = 2�DCT; where DCT =

CT(Target gene) � CT(GUS)) [20].

Fig. 1. Polymerase chain reaction (PCR) analysis of UPCI:SCC090

and CaSki DNA. Amplification of human papillomavirus (HPV)-16

genes L1, detected by consensus L1 primers, and E6, amplified using

type-specific HPV-16 primers as described in Section 2.

810 R.L. Ferris et al. / European Journal of Cancer 41 (2005) 807–815

2.7. RNA interference (siRNA) for HPV-16 E6

UPCI:SCC090 cells were plated in 24-well plates in

media, as above. At 50% confluence, cells were transfec-

ted with Oligofectamine (Invitrogen) following the man-

ufacturer�s protocol. siRNA specific for E6, or irrelevant(scrambled sequence) were ordered pre-duplexed and

ready for transfection from Dharmacon. Two HPV-16

E6 primers were mixed together using a total concentra-

tion of 60 pmol/well. siRNA target sequences are as fol-

lows: E6-1 5 0-AAGAGCUGAAACAACUAUAC-3 0,

E6-2 5 0-AACUGCGACGUGAGGUAUAUG-3 0, IR-

REL 5 0-AAGCACACACGUAGACAUUCG-3 0. Cells

were assayed for HPV16-E6 gene activity 48 h aftertransfection.

2.8. Determination of p53 expression and stability

Cells were treated with the indicated drugs and con-

centrations (described below) before lysis using l% Non-

idet P-40 (NP-40) and protease inhibitor cocktail

(Promega). Cisplatin was used at a final concentrationof 40 lM. Lactacystin [23] was used at a final concentra-

tion of 50 lM; MG-132 was used at a final concentra-

tion of 50 lM. Drug treatments were carried out for 6

h prior to cell lysis, sodium dodecyl sulphate-polyacryl-

amide gel electrophoresis (SDS-PAGE) electrophoresis

and Western blotting. Cellular extracts (5–10 ll totalprotein) were electrophoresed and immunoblotted using

anti-p53 DO-7 mAb (BD Pharmingen) and secondaryAb linked to horse radish peroxidase (HRP) or fluoro-

chromes (Amersham Pharmacia). Equal protein loading

was determined using the Pierce bicinchoninic acid

(BCA) protein quantitation reagent kit and confirmed

by blotting the polyvinylidene fluoride (PVDF) mem-

brane with anti-b-actin Ab (Sigma). The intensity of

protein bands was determined by densitometric analysis

of the immunoblots.

3. Results

3.1. UPCI:SCC090 cells contain integrated HPV-16

DNA

PCR analysis using the consensus M09/M11 primers

for the L1 gene showed the presence of a 450-bp band,

indicating the presence of HPV DNA in the UP-

CI:SCC090 cell line (Fig. 1). DNAs from CaSki (HPV-

16-positive) and C-33A (HPV-negative) were used as po-sitive and negative controls, respectively. In addition,

the G3PDH gene was amplified as a positive control

for all the cell lines. Since HPV-16 is the most common

type found in SCCHN, HPV-16-specific PCR reactions

were performed using DNA from this cell line as the

template. PCR analysis using primers for the E6 gene

showed that this cell line contained HPV-16 DNA

(Fig. 1). Primers specific for HPV-18 E6 gene failed to

amplify any DNA from the UPCI:SCC090 cell line(Fig. 1). To confirm that this cell line contains HPV-16

DNA, specific PCR primers were used to amplify por-

tions of the HPV-16 E2, E6, E7 and L1 genes. Auto-

mated DNA sequencing of 400–500 bp regions of the

above PCR products showed that the DNA corre-

sponded to that of HPV-16 (data not shown). Further-

more, the amplified DNA sequence resembled the

European E-G131G variant of HPV-16 [24]. These re-sults confirmed that the UPCI:SCC090 cell line contains

HPV-16 DNA.

We then carried out Southern blot analysis to deter-

mine the physical state of the HPV-16 DNA in the UP-

CI:SCC090 cell line. In these studies, we used DNA

from CaSki which contains tandem head to tail repeats

of integrated HPV-16 DNA [25] and the L-18 cell line

that contains episomal HPV-16 DNA [17] as controls.When the Southern blot was hybridised to a HPV-16

probe, uncut UPCI:SCC090 DNA and DNA cleaved

with BglII that does not cleave HPV-16 generated a sin-

gle, slow-migrating band similar to that observed with

the CaSki DNA (Fig. 2). By contrast, uncut L-18

DNA contained two major bands corresponding to the

supercoiled (SC) and open-circular (OC) forms of the

HPV-18 DNA which cross-hybridises to the HPV-16probe (Fig. 2). These results demonstrated that as is

the case with CaSki cells, the UPCI:SCC090 cell line

contains integrated HPV-16 DNA. When the UP-

CI:SCC090 DNA was treated with BamHI that cleaves

the HPV-16 DNA once, a major 7.8 kb band was ob-

served (Fig. 2). Since the HPV-16 DNA is present in

an integrated state in this cell line, the 7.8 kb band is pre-

sumably generated from the release of unit-length DNA

Fig. 2. Physical state of HPV-16 in the UPCI:SCC090 oral carcinoma

cell line, as determined by Southern blot analysis (sc, supercoiled).

Fig. 3. RT-PCR analysis of HPV-16 E2 and E6 transcripts in

UPCI:SCC090 cells.

R.L. Ferris et al. / European Journal of Cancer 41 (2005) 807–815 811

from tandemly-integrated HPV-16 sequences. CaSki

cells that are known to contain only integrated HPV-

16 DNA generated two major bands of similar intensity

of 7.9 and 6 kb when the DNA from this cell line was

treated with BamHI (Fig. 2). The 7.9 kb band corre-

sponds to the unit-length DNA released from tan-

demly-integrated HPV-16 sequences, while the smaller6 kb band corresponds to tandemly-integrated HPV-16

DNA containing deletions in the viral genome [25].

Treatment of the UPCI:SCC090 and CaSki DNA with

StuI and NcoI that also cleave the HPV-16 genome once

gave results similar to those observed with BamHI (data

not shown). As shown previously in Ref. [25], CaSki

cells also generated a faint 10.5 kb band that corre-

sponds to HPV-16 DNA containing a duplicated regionof the viral genome. Two fainter, larger than unit-length

bands observed with the UPCI:SCC090 sample (Fig. 2)

may also correspond to integrated HPV-16 genomes

containing duplicated viral sequences. Treatment of UP-

CI:SCC090 and CaSki DNA with KpnI that cleaves

HPV-16 twice generated two major bands of 4.4 and

3.4 kb (Fig. 2), consistent with the presence of tandemly

integrated HPV DNA. As a control for a cell line con-taining episomal HPV DNA, the L-18 DNA generated

an 7.8-kb band when cleaved with the single cutting

EcoRV enzyme and fragments of 6.8 and 1 kb (data

not shown) upon treatment with BamHI that cleaves

the HPV-18 DNA twice (Fig. 2). Taken together, our re-

sults suggest that the UPCI:SCC090 cell line contains

tandemly repeated copies of HPV-16 DNA integrated

into the chromosome.

3.2. Localisation of the region deleted in the HPV-16

LCR

To study the LCR of HPV-16 in the UPCI:SCC090

cell line, we amplified a 606-bp region of the virus by

PCR. Agarose gel analysis of the PCR product showedthat the reaction product was smaller than the expected

size obtained when wild-type HPV-16 DNA present in

the W12 cell line [26] was used as the template. To iden-

tify the site of deletion, the PCR-amplified LCR DNA

from UPCI:SCC090 was subjected to automated DNA

sequencing. These results showed that a 163-bp se-

quence of LCR (nucleotides 7658–7818) was deleted in

the HPV-16 genome present in the UPCI:SCC090 cellline.

3.3. HPV-16 genes are transcriptionally active in

UPCI:SCC090 cells

RT-PCR analysis was carried out to study the expres-

sion of HPV-16 E2 and E6 genes in the UPCI:SCC090

cell line. Before DNA synthesis, RNA was treated withDNaseI, amplification grade (Invitrogen), for 15 min at

room temperature to avoid DNA contamination. DNa-

seI was then inactivated by the addition of 25 mM

EDTA followed by incubation at 65 �C for 10 min. As

shown in Fig. 3, a single band predicted to be 554 bp

in size was obtained with the E2-specific primers. In

the presence of E6-specific primers, two expected bands

of 296 and 477 bp were observed (Fig. 3), correspondingto products of alternatively-spliced E6 mRNA [27,28].

Thus, UPCI:SCC090 expresses both the regulatory E2

812 R.L. Ferris et al. / European Journal of Cancer 41 (2005) 807–815

gene and the E6 oncogene of HPV-16, consistent with

the presence of tandemly integrated copies of the virus

in this cell line. Further, these results show that integra-

tion of the HPV-16 DNA in this cell line does not inter-

rupt the viral E2 gene.

3.4. Quantitative real-time PCR to quantitate relative

copies of HPV-16 E6 DNA

We performed qPCR for HPV-16 E6 and E7 gene to

quantify viral DNA copy numbers present in UP-

CI:SCC090 cells, compared with known high (CaSki)

and low (SiHa) copy number cervical carcinoma cells.

The cycle threshold (Ct) for UPCI:SCC090 cells was sig-nificantly lower than SiHa cells, which are known to

contain 1–2 copies of integrated HPV-16 genomic cop-

ies. Back-calculation using quantitatively amplified ser-

ies of microsatellite repeats as described in Ref. [22],

and developing a standard curve based on a serially di-

luted HPV-16 E6 encoding plasmid, enabled the relative

determination of HPV-16 copy numbers in UP-

CI:SCC090 cells. Thus, UPCI:SCC090 contains approx-imately 100–150 copies of HPV-16 DNA (Table 1),

similar to CaSki cells (known to contain several hundred

copies per host cellular genome).

3.5. Real-time quantitative reverse transcription PCR

demonstrates high expression of HPV-16 E6 and E7

oncogenes

We measured E6 and E7 mRNA levels in UP-

CI:SCC090 cells to evaluate the likelihood that HPV-

16 contributed to carcinogenesis in these cells. As shown

in Fig. 4, high-level E6 and E7 expression was observed.

Consistently higher (P < 0.005) levels of E6 and E7

expression were observed in UPCI:SCC090 cells com-

pared with CaSki and SiHa cells, indicating the likely

importance of these gene products in UPCI:SCC090carcinogenesis.

Fig. 4. Expression of HPV-16 E6 and E7 in UPCI:SCC090 cells by

quantitative reverse-transcription RT-PCR (qRT-PCR) relative to the

control gene, b-glucuronidase (GUS). *Relative to control QuMa

DNA amplication, using 2�DCT method.

3.6. Genotype of TP53 in UPCI:SCC090 cells

Although TP53 is mutated in some HPV-16 associ-

ated carcinomas, the mutation frequency is lower than

in non-HPV-associated tumours, and this correlates

with viral gene expression [29]. Consistent with this no-tion, we performed cDNA sequence analysis for the

TP53 gene in UPCI:SCC090 cells (data not shown). Di-

rect fluorogenic sequence analysis of both coding and

non-coding strands, with subsequent BLAST alignment

demonstrated that UPCI:SCC090 cells contain wild-

type TP53 in exons 2 through 11. Using differential

restriction digestion [30], we confirmed that UP-

CI:SCC090 cells are heterozygous for proline and argi-nine at codon 72.

3.7. Reversal of p53 degradation by anti-neoplastic and

anti-viral drug treatment

Based on the wild-type sequence of TP53 in UP-

CI:SCC090 cells, and E6-induced, ubiquitin-mediated

proteolysis of p53, we postulated that antineoplasticchemotherapeutic effects might be mediated through res-

toration of p53 expression. Fig. 5(a) shows an immuno-

blot of UPCI:SCC090 lysates, indicating that p53 is

essentially undetectable in these cells, likely due to the

enhanced ubiquitination and proteasomal degradation

induced by HPV-16 E6 protein associating with the

E6-AP (ubiquitin E3 ligase). After cisplatin treatment

(40 lM for 6 h), reversal of this E6 effect and p53 expres-sion was clearly observed in UPCI:SCC090 and CaSki

cells (data not shown). A similar effect was also seen

with proteasome inhibitors lactacystin [23] and leuci-

nal–leucinal–norleucinal (referred to as LLnL or MG-

132).

3.8. siRNA knockdown of HPV-16 E6 results in

accumulation of p53

To clarify the carcinogenic effect of HPV-16 E6 on

p53 expression, UPCI:SCC090 cells were transfected

with siRNA against HPV-16 E6. E6 RNA levels were re-

duced up to 70% as shown by qRT-PCR (data not

shown). We found that repression of HPV-16 E6 expres-

sion led to the accumulation of p53 in these cells (Fig.

5(b)). The level of p53 accumulation observed in siR-NA-treated UPCI:SCC090 cells was lower than that

seen after treatment with potent proteasome inhibitors

(Fig. 5(a)), likely due to incomplete knockdown of E6

expression by siRNA.

4. Discussion

Epidemiological correlation between mucosatropic

HPV infection and a subgroup of SCCHN is most com-

Fig. 5. Recovery of p53 expression by treatment of UPCI:SCC090

cells with the chemotherapeutic agent cisplatin or by proteasome

inhibitors (a) or by downregulation of HPV-16 E6 expression by RNA

interference (b). (a) Recovery of p53 from HPV-l6-induced degrada-

tion. Western blot analysis of the p53 protein in UPCI:SCC090 and

CaSki cells using DO-7 anti-p53 monoclonal antibodies after treat-

ment with cisplatin (40 lM for 6 h; lanes 2 and 6) and after treatment

with specific proteasome inhibitors lactacystin (10 lM; lanes 3 and 7)

or MG-132 (50 lM, lanes 4 and 8). (b) Western blot analysis of the p53

protein after transient transfection of UPCI:SCC090 cells with a 21-

mer oligonucleotides designed for short hairpin RNA interference of

the E6 gene (siRNA), see Section 2 (1FN-c, interferon-c).

R.L. Ferris et al. / European Journal of Cancer 41 (2005) 807–815 813

pelling for the oropharyngeal site (tonsil or base of ton-

gue). In general, tobacco exposure is inversely associated

with the presence of HPV-infection in SCCHN, but

these features are not absolute. The determinants of

each factor in the aetiology of SCCHN has been difficult

to identify due to a lack of natural in vitro cellular mod-

els of this disease. Cell lines such as we have described in

this report are critical to our understanding of this dis-ease and comparing features between HPV-associated

SCCHN and cervical carcinoma are limited. Reports

of propagation of naturally occurring, HPV-trans-

formed SCCHN cell lines have been infrequent, with

few cell lines maintaining their HPV DNA [31,32]. Epi-

somal HPV DNA, by itself, is generally expected to be

an unlikely aetiological factor for SCCHN, due to the

low-level expression of the transforming E6 and E7 pro-teins. However, the presence of mutations in HPV LCR

could result in an increased expression of these onco-

genes [25,31,33–35] and this may lead to cellular trans-

formation. Integrated viral DNA that leads to high

levels of E6/E7 expression is much more likely to be

present in naturally transformed, HPV-16-infected

SCCHN cell lines. Because a causal association of

HPV-induced carcinogenesis is uncertain, the need forin vitro cellular models is paramount to enable the char-

acterisation of HPV-associated SCCHN.

Recently, it has been shown by fluorescent in situ

hybridisation (FISH) and restriction-site PCR that the

UPCI:SCC090 cell line contains integrated HPV-16

DNA in chromosomes 3, 6, 9q and 13q [33]. Further-

more, the integration sites were found to be generally lo-

cated in common fragile sites. Our quantitative PCRstudies demonstrate that the HPV-16 DNA is present

at approximately 100–150 copies per cell in UP-

CI:SCC090 (Table 1). Southern blot analysis of uncut

UPCI:SCC090 DNA with an HPV-specific probe

showed the presence of a high molecular weight DNA,

but no signal corresponding to the SC or OC forms of

HPV-16 DNA was observed (Fig. 2). Since treatment

of UPCI:SCC090 DNA with BamHI that cleavesHPV-16 DNA only once generated a single band of

7.8 kb (Fig. 2), our results suggest that the viral genome

is integrated in tandem copies in a head to tail orienta-

tion in this cell line. Such an arrangement of integrated

sequences would release unit length HPV-16 DNA upon

treatment with a single-cutting enzyme. This interpreta-

tion is consistent with the observations with the CaSki

cell line which also generates an 8 kb band upon treat-ment with BamHI (Fig. 2). However, CaSki cells also

generated a second major band of approximately 6 kb,

which represents tandem integration of a second, deleted

form of HPV-16 DNA in this cell line [25]. Deletions

and rearrangements of HPV DNA are frequently ob-

served in cervical carcinoma cell lines containing inte-

grated HPV DNA [33–35]. We used DNA from the L-

18 cell line containing episomal copies of HPV-18DNA as a control. The results obtained with the L-18

DNA were consistent with the presence of episomal

HPV-18 DNA in this cell line (Fig. 2).

UPCI:SCC090 cells contain integrated HPV-16 DNA

and express high levels of E6, which is important for cel-

lular transformation [36,37]. These observations are

consistent with a causative role for HPV-16 in the carci-

nogenesis of SCCHN. Interestingly, UPCI:SCC090 alsoexpressed the viral E2 gene (Fig. 3). This suggests that,

as is the case in CaSki cells, the E2 gene is not disrupted

or deleted in the UPCI:SCC090 cell line. Previous stud-

ies have identified several mutations in the LCR of

HPVs in cases of cervical carcinoma, especially those

in which the HPV genome is present in an episomal

form [1,34]. PCR amplification of the HPV-16 LCR in

the UPCI:SCC090 cell line followed by DNA sequenc-ing showed that a 163-bp sequence (nucleotides 7656–

7818) was deleted in the HPV-16 genome present in

the UPCI:SCC090 cell line. This region includes a por-

tion of the HPV-16 enhancer sequence, two YY1 bind-

ing sites and one NF1 binding site [34,35]. Regions

within the HPV LCR are known to negatively regulate

HPV-16 E6 and E7 expression. For example, binding

of the YY1 factor to its binding site has been suggestedto interfere with the formation of the transcription initi-

ation complex at the P97 promoter for the E6/E7 genes

814 R.L. Ferris et al. / European Journal of Cancer 41 (2005) 807–815

[35]. Since the HPV-16 genome in the UPCI:SCC090 cell

line contains a deletion of the binding sites for the YY1

and NFI proteins, our results suggest that this may re-

sult in the upregulation of E6/E7 expression, even in

the presence of viral E2 protein. The additional se-

quences missing in the HPV-16 LCR in UPCI:SCC090cells may also play a negative role in E6/E7 oncogene

expression. Integrated copies of HPV-16 are also present

in the tumour from which the UPCI:SCC090 cell line

was derived (data not shown), suggesting that expres-

sion of the E6/E7 oncogenes from the integrated state

may have contributed to the development of the tumour

from which this cell line was derived.

Phenotypic and genotypic analysis shows that UP-CI:SCC090 represents an interesting and valid model

of HPV-16-associated SCCHN. It is derived from a de

novo oropharyngeal tumour, a site with roughly 50%

prevalence of HPV-16 DNA in numerous studies

[38,39]. This cell line contains wild-type p53 which is

rapidly degraded in untreated cells, but is recoverable

by the action of specific proteasome inhibitors and E6

knockdown by siRNA (Fig. 5), consistent with the func-tional p53 inactivation by the HPV-16 E6 protein. Bio-

logically active concentrations of cisplatin, a commonly

used chemotherapeutic drug in SCCHN, showed similar

recovery of p53 expression (Fig. 5(a)). Although cervical

cancer may not be as sensitive to this drug as SCCHN,

this fact raises the possibility that these diseases

although generally associated with the same viral sub-

types, may develop through distinct aetiological mecha-nisms. Such issues may now be tested in vitro comparing

features of cell lines derived from individuals with each

disease.

Because of the mounting molecular epidemiological

data in favour of the presence an HPV-16-associated

subtype of SCCHN, particularly in the oropharynx,

the UPCI:SCC090 cell line promises to be quite valuable

as an in vitro model system to test similarities betweenHPV-associated SCCHN, HPV-negative SCCHN, and

HPV-induced cervical carcinoma. Further studies are

expected to clarify the aetiological role of HPV-16 in a

clinically significant subset of SCCHN.

Conflict of interest statement

None declared.

Acknowledgements

This work was supported by the Hillman Founda-

tion, American Head and Neck Society, The American

Academy of Otolaryngology/Head and Neck Surgery,the Stout Family fund for Clinical Research, a pilot

grant from the Oral Cancer Center at the University

of Pittsburgh and the University of Pittsburgh Eye

and Ear Foundation.

References

1. zur Hausen H. Papillomaviruses and cancer: from basic studies to

clinical application. Nat Rev Cancer 2002, 2(5), 342–350.

2. Herrero R, Castellsague X, Pawlita M, et al. Human papilloma-

virus and oral cancer: The International Agency for Research on

Cancer multicenter study. J Natl Cancer Inst 2003, 95(23),

1772–1783.

3. Scheffner M, Romanczuk H, Munger K, et al. Functions of

human papillomavirus proteins. Curr Top Microbiol Immunol

1994, 186, 83–99.

4. Huibregtse JM, Scheffner M, Howley PM. E6-AP directs the HPV

E6-dependent inactivation of p53 and is representative of a family

of structurally and functionally related proteins. Cold Spring Harb

Symp Quant Biol 1994, 59, 237–245.

5. Braakhuis BJ, Snijders PJ, Keune WJ, et al. Genetic patterns in

head and neck cancers that contain or lack transcriptionally

active human papillomavirus. J Natl Cancer Inst 2004, 96(13),

998–1006.

6. Park JS, Hwang ES, Park SN, et al. Physical status and

expression of HPV genes in cervical cancers. Gynecol Oncol

1997, 65(l), 121–129.

7. Ke LD, Adler-Storthz K, Mitchell MF, et al. Expression of

human papillomavirus E7 mRNA in human oral and cervical

neoplasia and cell lines. Oral Oncol 1999, 35(4), 415–420.

8. Greenlee RT, Hill-Harmon MB, Murray T, et al. Cancer statis-

tics, 2001. CA Cancer J Clin 2001, 51(l), 15–36.

9. Bercovich JA, Centeno CR, Aguilar OG, et al. Presence and

integration of human papillomavirus type 6 in a tonsillar

carcinoma. J Gen Virol 1991, 72(Pt 10), 2569–2572.

10. Scheurlen W, Stremlau A, Gissmann L, et al. Rearranged HPV 16

molecules in an anal and in a laryngeal carcinoma. Int J Cancer

1986, 38(5), 671–676.

11. Steenbergen RD, Hermsen MA, Walboomers JM, et al. Inte-

grated human papillomavirus type 16 and loss of heterozygosity at

11q22 and 18q21 in an oral carcinoma and its derivative cell line.

Cancer Res 1995, 55(22), 5465–5471.

12. Venuti A, Manni V, Morello R, et al. Physical state and

expression of human papillomavirus in laryngeal carcinoma

and surrounding normal mucosa. J Med Virol 2000, 60(4),

396–402.

13. Maitland NJ, Cox MF, Lynas C, et al. Detection of human

papillomavirus DNA in biopsies of human oral tissue. Brit J

Cancer 1987, 56(3), 245–250.

14. Vernon SD, Unger ER, Miller DL, et al. Association of human

papillomavirus type 16 integration in the E2 gene with poor

disease-free survival from cervical cancer. Int J Cancer 1997, 74(l),

50–56.

15. Yoo GH, Washington J, Oliver J, et al. The effects of exogenous

p53 overexpression on HPV-immortalized and carcinogen trans-

formed oral keratinocytes. Cancer 2002, 94(1), 159–166.

16. Heo DS, Snyderman C, Gollin SM, et al. Biology, cytogenetics,

and sensitivity to immunological effector cells of new head and

neck squamous cell carcinoma lines. Cancer Res 1989, 49(18),

5167–5175.

17. Frattini MG, Lim HB, Doorbar J, et al. Induction of human

papillomavirus type 18 late gene expression and genomic ampli-

fication in organotypic cultures from transfected DNA templates.

J Virol 1997, 71(9), 7068–7072.

18. Maniatis T, Fritsch EF, Sambrook J. Molecular cloning : a

laboratory manual. 2nd ed. Cold Spring Harbor, NY, Cold

Spring Harbor Laboratory, 1989.

R.L. Ferris et al. / European Journal of Cancer 41 (2005) 807–815 815

19. Manos MM, Ting Y, Wright DK, et al. Use of polymerase chain

reaction amplification for the detection of genital human papil-

lomavirusesCancer cells. pp. 209–214.

20. Wang J, Xi L, Hunt JL, et al. Expression pattern of chemokine

receptor 6 (CCR6) and CCR7 in squamous cell carcinoma of the

head and neck identifies a novel metastatic phenotype. Cancer Res

2004, 64(5), 1861–1866.

21. Wang-Johanning F, Lu DW, Wang Y, et al. Quantitation of

human papillomavirus 16 E6 and E7 DNA and RNA in residual

material from ThinPrep Papanicolaou tests using real-time poly-

merase chain reaction analysis. Cancer 2002, 94(8), 2199–2210.

22. Tassone F, Hagerman RJ, Taylor AK, et al. Elevated levels of

FMR1 mRNA in carrier males: a new mechanism of involvement

in the fragile-X syndrome. Am J Hum Genet 2000, 66(1), 6–15.

23. Fenteany G, Standaert RF, Reichard GA, et al. A beta-lactone

related to lactacystin induces neurite outgrowth in a neuroblas-

toma cell line and inhibits cell cycle progression in an osteosar-

coma cell line. Proc Natl Acad Sci USA 1994, 91(8), 3358–3362.

24. Yamada TMM, Peto J, Greer CE, et al. Human papillomavirus

type 16 sequence variation in cervical cancers: a worldwide

perspective. J Virol 1997, 71(3), 2463–2472.

25. Meissner JD. Nucleotide sequences and further characterization

of human papillomavirus DNA present in the CaSki, SiHa and

HeLa cervical carcinoma cell lines. J Gen Virol 1999, 80(Pt 7),

1725–1733.

26. Stanley MA, Browne HM, Appleby M, et al. Properties of a non-

tumorigenic human cervical keratinocyte cell line. Int J Cancer

1989, 43(4), 672–676.

27. Fuchs PG, Pfister H. Transcription of papillomavirus genomes.

Intervirology 1994, 37(3–4), 159–167.

28. Mansur CP, Androphy EJ. Cellular transformation by papillo-

mavirus oncoproteins. Biochim Biophys Acta 1993, 1155(3),

323–345.

29. Dai M, Clifford GM, le Calvez F, et al. Human papillomavirus

type 16 and TP53 mutation in oral cancer: matched analysis of the

IARC multicenter study. Cancer Res 2004, 468–471.

30. Shen H, Zheng Y, Sturgis EM, et al. P53 codon 72 polymorphism

and risk of squamous cell carcinoma of the head and neck: a case-

control study. Cancer Lett 2002, 183(2), 123–130.

31. Shillitoe EJ, Noonan S. Strength and specificity of different gene

promoters in oral cancer cells. Oral Oncol 2000, 36(2), 214–220.

32. Sacks PG. Cell, tissue and organ culture as in vitro models to

study the biology of squamous cell carcinomas of the head and

neck. Cancer Metast Rev 1996, 15(1), 27–51.

33. Ragin CC, Reshmi SC, Gollin SM. Mapping and analysis of

HPV16 integration sites in a head and neck cancer cell line. Int J

Cancer 2004, 110(5), 701–709.

34. Rose BR, Thompson CH, Zhang J, et al. Sequence variation in

the upstream regulatory region of HPV 18 isolates from cervical

cancers. Gynecol Oncol 1997, 66(2), 282–289.

35. O�Connor MJ, Tan SH, Tan CH, et al. YY1 represses human

papillomavirus type 16 transcription by quenching AP-1 activity. J

Virol 1996, 70(10), 6529–6539.

36. Banks L, Crawford L. Analysis of human papillomavirus type 16

polypeptides in transformed primary cells. Virology 1988, 165(l),

326–328.

37. Crook T, Morgenstern JP, Crawford L, et al. Continued expres-

sion of HPV-16 E7 protein is required for maintenance of the

transformed phenotype of cells co-transformed by HPV-16 plus

EJ-ras. EMBO J 1989, 8(2), 513–519.

38. Mork J, Lie AK, Glattre E, et al. Human papillomavirus

infection as a risk factor for squamous-cell carcinoma of the head

and neck. N Engl J Med 2001, 344(15), 1125–1131.

39. Gillison ML, Koch WM, Capone RB, et al. Evidence for a causal

association between human papillomavirus and a subset of head

and neck cancers. J Natl Cancer Inst 2000, 92(9), 709–720.